The chemistry of thiocarbonic acids and their salts has been studied in some detail, as indicated by O'Donoghue and Kahan, Journal of the Chemical Society, Vol. 89(II), pages 1812-1818 (1906); Yeoman, Journal of the Chemical Society, Vol. 119, pages 38-54 (1921); Mills and Robinson, Journal of the Chemical Society, Vol. 128(II), pages 2326-2332 (1928) and by Stone et al. in U.S. Pat. No. 2,893,835, dated Jul. 7, 1959.
According to O'Donoghue and Kahan, as far back as 1826 derivatives of thiocarbonic acid were prepared by Berzelius, who reacted aqueous solutions of hydrosulfides with carbon disulfide to give unstable solutions which yielded unstable crystalline salts in accordance with the following reaction: EQU 2 KSH+CS.sub.2 =&gt;K.sub.2 CS.sub.3 +H.sub.2 S (1)
Other thiocarbonates were prepared and further characterized by O'Donoghue and Kahan. Their paper, at page 1818, reports the formation of ammonium thiocarbonate by reacting liquid ammonia with cold alcoholic thiocarbonic acid prepared by dropping a solution of calcium thiocarbonate into concentrated hydrochloric acid to produce free thiocarbonic acid (H.sub.2 CS.sub.3). The calcium thiocarbonate utilized by the authors is described as a double salt, including the calcium cation in combination with both the hydroxide and the trithiocarbonate anions. In addition to free thiocarbonic acid, other compounds prepared by O'Donoghue and Kahan included the sodium, potassium, zinc and lead salts. However, regardless of which of these salts were prepared, a common characteristic was their relative instability, with the prepared compounds breaking down and releasing carbon disulfide and hydrogen sulfide and/or a metal sulfide, often in a matter of minutes.
The noted paper by Yeoman reports a further study of thiocarbonates (called trithiocarbonates therein) and also reports the preparation and properties of perthiocarbonates (or tetrathiocarbonates), derivatives of tetrathiocarbonic acid (H.sub.2 CS.sub.4). Yeoman reports on methods of preparing the ammonium, alkali metal and alkaline earth metal salts of these acid species. For example, Yeoman prepared ammonium trithiocarbonate by saturating an alcoholic ammonia solution with hydrogen sulfide and then adding carbon disulfide to precipitate the product salt. Ammonium perthiocarbonate was prepared in a similar manner, except that after reacting the ammonia and hydrogen sulfide, elemental sulfur was added to form the disulfide, (NH.sub.4).sub.2 S.sub.2 ; adding carbon disulfide immediately precipitated the product.
Yeoman states that solutions of both ammonium trithiocarbonate and perthiocarbonate are very unstable due both to decomposition to form thiocyanate as a product, and to complete dissociation back into ammonia, hydrogen sulfide and carbon disulfide.
Considerable explanation is provided concerning the stability of thiocarbonates, as exemplified by sodium trithiocarbonate and perthiocarbonate. Sodium trithiocarbonate solutions in water are said to remain stable only if oxygen and carbon dioxide are rigidly excluded; the presence of oxygen causes decomposition to form carbon disulfide and thiosulfates, while carbon dioxide decomposes the solution to form a carbonate, elemental sulfur, carbon disulfide and hydrogen sulfide. Potassium trithiocarbonate behaves similarly, according to Yeoman.
Yeoman also attempted to prepare and characterize the stability of thiocarbonate salts of four of the alkaline earth metals. Yeoman was unable to prepare a pure calcium tri- or tetrathiocarbonate, but did observe that the double salt of calcium trithiocarbonate which he prepared was more stable (probably because it was less hygroscopic) than the sodium or potassium thiocarbonates. The barium salt of tetrathiocarbonic acid could not be isolated, although Yeoman believed it existed in solution. Solid barium trithiocarbonate could not be isolated, although it was alleged to behave like sodium trithiocarbonate when dissolved in water. The preparation of aqueous solutions of the tri- and tetrathiocarbonates of magnesium and strontium was alleged, but the magnesium thiocarbonates were not isolated.
The previously noted paper by Mills and Robinson shows the preparation of ammonium thiocarbonate by digesting ammonium pentasulfide (obtained by suspending sulfur in aqueous ammonia, then saturating with hydrogen sulfide) with carbon disulfide. A crystalline residue from the reaction was found to be ammonium perthiocarbonate. The authors prepared a "better" ammonium perthiocarbonate product, however, by extracting the ammonium pentasulfide with carbon disulfide in a Soxhlet apparatus.
Stone et al. disclose several methods for preparing solid ammonium, alkali and alkaline earth metal salts of tri- and tetraperoxythiocarbonates, hereinafter referred to simply as "tetrathiocarbonates." One such method involves the solution of an active metal such as sodium in anhydrous ethanol to form an ethoxide which, in turn, is reacted with hydrogen sulfide and carbon disulfide to form sodium trithiocarbonate. They report, however, that the trithiocarbonates tend to be quite soluble in ethanol, and if it is desired to recover the solid material from the solution, it is necessary to treat the reaction mixture with a "displacing agent" such as ether, in which case the thiocarbonates frequently separate, not as solids, but as difficultly crystallizable oils which appear to be saturated aqueous solutions of the trithiocarbonate salt. Consequently, such a procedure is not considered feasible for use on a commercial scale. Similar problems were reported with tetrathiocarbonate salts, which were prepared using procedures analogous to those for the trithiocarbonates.
These problems were reportedly solved by carrying out the preparation reaction in a medium which is composed of a major part of a nonsolvent for the reaction components and a minor proportion of a liquid which is miscible with the nonsolvent and which is a solvent, to a measurable degree, for inorganic sulfides. The preferred nonsolvents used were relatively low boiling hydrocarbon materials such as hexane, cyclohexane and benzene. The second solvent was preferably ethanol, isopropanol or dioxane.
Basic physical and chemical properties of these materials and a number of methods for making them are summarized in considerable detail, starting at page 154 in "Carbon Sulfides and their Inorganic and Complex Chemistry" by G. Gattow and W. Behrendt, Volume 2 of "Topics in Sulfur Chemistry", A. Senning, Editor, George Thieme Publishers, Stuttgart, 1977.
What is needed is a process for the manufacture of salts of tetrathiocarbonic acid which is less cumbersome than the processes previously used. Such process should be capable of providing aqueous solutions of tetrathiocarbonates on a commercial scale. The present invention provides such a process.